Page 42 - ITU Journal - ICT Discoveries - Volume 1, No. 2, December 2018 - Second special issue on Data for Good

ITU JOURNAL: ICT Discoveries, Vol. 1(2), December 2018

temperature, conductivity, pH and dissolved oxygen, The goal is to develop a solution for easy
since advanced sensors measuring for example interpretation of sensor data, where changes in
compounds are too costly. water quality can be observed and potential causes
of these changes are suggested by the system. One
ICT companies, such as Nokia, Microsoft and example could be the indication of algae bloom,
Huawei are investigating how water quality where the individual sensor parameters show an
monitoring can be utilized and fit into their IoT increased temperature and conductivity together
offerings for smart sustainable cities. However, to with lowered oxygen content in the water. The
our knowledge, no major ICT company has any introduction of such AI functionalities will be a
comprehensive offering as of today. During 2016, differentiator compared with exisitng solutions,
Ericsson initiated a project to develop a and take digitalized water quality monitoring to a
comprehensive cloud-based water quality new level. This final phase was initiated during the
monitoring solution, based on a massive IoT spring of 2018 and will concentrate on water
approach. The project, developed in Stockholm, contamination models, development of algorithms
Sweden, was based on a previous proof-of-concept for the AI functionality and integration to a
in the US [17] and is a collaboration between comprehensive product for smart cities.
Ericsson, the city of Stockholm, academia and other
companies in the Stockholm region [18]. 4. DISCUSSION

4.1 The benefits and challenges of environ-
mental data

As the previous section showed, there is a strong
Fig. 3 – Technical architecture of a typical digital real-time focus on developing solutions for environmental
environmental monitoring solution.
monitoring in general and water quality monitoring
specifically, but how should these solutions be used?
The goal of the project, which is still ongoing, has The following section will elaborate on the benefits
been to to develop a comprehensive IoT-based as well as the challenges for digital environmental
digital water monitoring solution. Based on the first monitoring solutions. Furthermore, a number of
phase that was finalized in 2017, a commercial obstacles need to be solved to fully take advantage
solution was launched during 2018 covering water of these technologies.
and air quality as well as noise measurements for
smart cities [19]. During the first phase of the Ericsson is expecting that ICT will be a driver for
project, sensors were deployed in Lake Mälaren, in enabling the potential of the smart sustainable cities
Stockholm, measuring different basic parameters of the future, and to accelerate the achievement of
(pH, temperature, conductivity, dissolved oxygen the SDGs [20]. IoT, 5G and artificial intelligence will
and redox potential). The sensors were potentially be powerful enablers [21] and
communicating over the LTE network with digitalized real-time environmental data can be
Ericsson’s cloud-based IoT platform (see Fig. 3). efficient tools for urban planning and management
in the sustainable cities of the future. Urban city
Lessons learned from this first phase was that there data can be collected from various sources, such as
is a discrepancy between sensor data and reality. In traffic, homes, buildings, air quality and water
practice this means that it is very difficult for an end quality as well as weather data in general and rain
user to understand and interpret sensor parameter data specifically. Individually, all these systems
data and evaluate if there are any changes in the provide valuable data for the city planner and the
water quality induced by pollution or conta- citizens, but the true value of such data is obtained
mination. Hence, the next phase of the project will when all the systems work together combining the
be to utilize multivariate analysis and artificial aggregated data.
intelligence (AI) to understand the water quality
parameters and identify potential pollution and 4.1.1 ICT as an enabler of environmental
pathogenic contamination of the entire water monitoring
supply of a city.
Digital environmental monitoring, such as rain
monitoring and water quality solutions, is a

20 © International Telecommunication Union, 2018